Smart surveillance and the caesarean section rate: Have they cracked the code?

A recently published paper from Taiwan has caught my eye (Cheng, et al., 2023). Researchers don’t often find new technology can reduce the caesarean section rate, but that’s what this group have done. Let’s dig in and see what it was all about. (The paper is open access so feel free to grab a copy and read along with me).

What was the new tech?

The authors described a newly developed software system, called iWard. The software incorporated a maternal early warning system (a way of charting a woman’s vital signs with alerts when they fall outside expected ranges, with prompts about what to do about that), a fetal heart rate interpretation algorithm, a partogram (a graph showing changes in cervical dilation, and sometimes also the descent of the fetal head, over time), and “intrapartum care bundles”. I take these to mean guidelines about labour management. The software provided a “dashboard” at the “nursing station” and could send automatic alerts to the obstetric team when something fell outside expected ranges.

What was missing from the description was exactly what was going on with fetal heart rate monitoring during labour. I can’t tell whether the women in this study were using CTG monitoring, or intermittent auscultation, or a mix of the two approaches. My impression is they were all using CTG monitoring – Figure 4 describes data coming from “fetal monitors” and this reads as CTG generated data to me. (Once again we see the idea that CTG monitoring is fetal monitoring and intermittent auscultation is “not monitoring”.)

Who made up the study population?

Two groups of women were identified: those giving birth in a six month period immediately prior to the implementation of the new software, and those giving birth in a six month period that started two months after implementation. This gave everyone time to settle into the new system. All up, 3648 women were included, 1760 in the “before” period and 1888 in the “after” period. These groups were further divided into women with no previous caesarean section, and another group of women giving birth for the first time (nulliparous), at term, with one baby (singleton), who was head down (vertex – making this group NTSV, this is Robson category 1 & 2 for those who speak that language). The numbers here, as expected, were smaller – 381 before, and 433 after implementation of the software.

We are given no further information about the women – whether they had pregnancy complications or other health issues, whether their labour onset was spontaneous or not, or about the use of epidural analgesia, for example. All these things might also impact on mode of birth, and ideally should be controlled for in analyses. The authors also provide no information about instrumental birth rates. Did they go up as the caesarean section rate fell? So there are some notable gaps in this research.

What did they find?

Across the total population there was a small but statistically significant fall in the caesarean rate: from 46.8% to 42.5%, a risk ratio of 0.91 (95% confidence interval 0.88 – 0.95). For women without a history of previous caesarean section the size of the fall was similar: from 37.0% to 33.0% (risk ratio 0.89, 95% confidence interval 0.81 – 0.98). For women in the NTSV category, the fall was larger: from 31.0% to 23.3% (risk ratio 0.75, 95% confidence interval 0.71 – 0.80).

The authors provided information about changes in the use of caesarean section for specific reasons in figure 2. Of note here:

there was essentially no difference in the use of caesarean section for abnormal heart rate patterns in the NTSV population (9.4% before, 9.5% after),
there was a small reduction in the use of caesarean section for abnormal heart rate patterns for women with no previous caesarean section (8.8% before, 7.5% after),
the biggest difference was in the use of caesarean section for “labour arrest” (for the NTSV group it fell from 20.7% to 12.9%, and in the no previous caesarean section group from 11.7% to 8.4%),
There were no differences in the use of caesarean section for other indications than these.

Something interesting happened in the NTSV group with respect to gestational age at birth – pregnancies where statistically significantly longer after implementation. The difference was only a few days (38.5 weeks before compared 38.8 weeks after). Does this reflect a lower rate of labour induction in the later time period perhaps? That might also explain the fall in caesarean section rates. It is quite an early mean gestation and implies a high rate of labour induction.

There were also strongly significant increases in the duration of labour in the NTSV population. Mean labour duration rose from 9.2 hours to 12.3 hours. This could represent more tolerance for slower labour progress explaining the fall in the use of caesarean section for “labour arrest”.

No difference in any measured perinatal outcome (Apgar score, birth weight, admission to the nursery, meconium aspiration, chorioamnionitis, shoulder dystocia) or in a composite measure of adverse outcome, was seen. This provides reassurance the lower caesarean section rate was not placing the fetus at higher risk.

Should this surveillance system be introduced in other places?

I have unanswered questions. It would be great to see this applied in a large before and after study, where they gathered data about potential confounders (like labour induction) and controlled for these. I’d also like to know more about what the system actually includes and does not include (is CTG use the only approach to fetal heart rate monitoring the system can work with?). As I found with central fetal monitoring (Small, et al., 2021), introducing complex software into complex social settings (like birth suites) can have unintended consequences. It would be good to have research running in parallel with the introduction of a system like this to understand the impact it has on how maternity professionals provide care and how women experience that care.

The authors were aware of the potential for adverse impacts, writing:

implementing such a smart system should go beyond too little, too late, too much, too soon scenario and pursue a pathway towards evidence-based, respectful maternity care.
p. 126

I wholeheartedly agree!

The biggest impact here seems to be providing women time to complete labour without offering caesarean section. The authors explained the algorithm in the software was based on the World Health Organization Labour Care Guide. I wonder whether a less expensive low-tech option to simply shift to this guideline might have achieved the same result. The authors noted the increased medical cost associated with longer labour duration and have suggested a cost-benefit analysis. While I agree there is value in this, I would hate to find the priorities for women’s birth care are dictated by the desire to achieve cost effectiveness for hospitals rather than by women’s goals for their births.

Want to stay up-to-date with the latest research and course offers? Our monthly newsletter is here to keep you in the loop.

By subscribing to the newsletter, you’ll gain exclusive access to:

Exciting Announcements: Be the first to know about upcoming courses. Stay ahead of the curve and grab your spot before anyone else!
Exclusive Offers and Discounts: As a valued subscriber, you’ll receive special discounts and offers on courses. Don’t miss the chance to save money while investing in your professional growth.

Join the growing community of birth folks by signing up for our newsletter today!

Subscribe to the newsletter

References

Cheng, P. J., Cheng, Y. H., Shaw, S. S. W., & Jang, H. C. (2023, Jul 11). Reducing primary cesarean delivery rate through implementation of a smart intrapartum surveillance system. NPJ Digital Medicine, 6(1), 126. https://doi.org/10.1038/s41746-023-00867-y

Small, K., Sidebotham, M., Gamble, J., & Fenwick, J. (2021, Jun 24). “My whole room went into chaos because of that thing in the corner”: Unintended consequences of a central fetal monitoring system. Midwifery, 102, 103074. https://doi.org/10.1016/j.midw.2021.103074